The satellite HIPPARCOS, developed by the European Space Agency was launched from Kourou in French Guiana in August 1989. It was the first ever 'astrometric' satellite, designed to measure very accurately the angular positions of about 120,000 stars and the way these positions change with time. In the UK, astronomers at the Royal Greenwich Observatory (RGO) have been actively involved at all stages from the planning of the mission to the present analysis of the data. In the construction phase, British Aerospace were subcontracted to the prime contractors MATRA (F) to provide the Electrical Power and Attitude Measurement Control subsystems and LOGICA to provide on-board software.

Angular positions can be used to determine the distances to nearby stars and to study the motions of stars in our galaxy. Distances are of fundamental importance in astrophysics, and the only direct method of measuring them is by triangulation. As the Earth moves around the sun, a minute steady change in the apparent position of a nearby star against the background of more distant stars may be detected. This annual change is called parallax. Using HIPPARCOS the angular measurements possible are comparable to measuring the height of a man standing on the moon.

The accuracy of ground based observations is limited by the blurring of stellar images by the Earth's atmosphere, relating observations made in the northern and southern hemispheres and accounting for gravitational distortions in an Earth-bound telescope as it is moved around the sky. Accurate positions measured by HIPPARCOS will push the frontier of direct distance measurement to include a volume 125 times larger than before.

The telescope on HIPPARCOS has a main mirror of diameter 29 cm. As the satellite slowly scans the sky, stellar images from two fields of view, separated by 58°, cross behind a modulating grid at the focus of the telescope enabling accurate timing measurements of the crossing to be made. These relative timings are converted to relative positions in the data reduction process.

The data from HIPPARCOS are being independently analysed by two consortia of scientists working at institutions throughout Europe. The RGO team, led by Dr Floor van Leeuwen, is part of the `Northern Data Analysis Consortium' with other teams in Denmark and Sweden. No results will be available until the end of the mission, when the vast network of relative angular positions produced by the two independent groups is put together to produce a self consistent map of the sky, tied into an 'inertial' reference frame defined by distant galaxies.

The HIPPARCOS satellite carried out its measurements in a highly elliptical 10-hour orbit, ranging between 500 km and 36,000 km above the Earth's surface, resulting from the non-functioning of the satellite's apogee boost motor shortly after launch. A redesign of the on-board attitude control system, and the addition of two more ground stations into the control network, nevertheless allowed operation of the satellite with close to full efficiency.

The orbit, however, subjected the satellite to very significant levels of high energy electron and proton radiation, much higher than those expected in its intended geostationary orbit.

Mission accomplished

After more than three years of efficient and successful operations, communications with HIPPARCOS were terminated on 15 August 1993. Targetted for an operational lifetime of two and a half years, more than three years of high quality star measurements were eventually accumulated, and all of the original scientific goals of the mission have been fully accomplished.

During the last few months of its life, as the high radiation environment to which the satellite was exposed took its toll on the on-board systems, it was operated with only two of the three gyroscopes normally required for such a satellite, following an ambitious redesign of the on-board and on-ground systems. Plans were in hand to operate the satellite without gyroscopes at all, and the first such 'gyro-less' data had been acquired, when communication failure with the on-board computers on 24 June put an end to the relentless flow of data that have been sent down from the satellite since launch. Further attempts to continue operations proved unsuccessful, and after a short series of sub-system tests, operations were terminated four years and one week after launch.

An enormous wealth of scientific data was gathered by HIPPARCOS. Even though data analysis by the scientific teams involved in the programme is not yet complete, it is clear that the mission has been an overwhelming success.

Extremely accurate positions for more than one hundred thousand stars, precise distance measurements (in most cases, for the first time), and accurate determinations of the stars' velocity through space have been derived. The resulting HIPPARCOS Star Catalogue, expected to be completed in 1996, will be of unprecedented accuracy, achieving results some 10-100 times more accurate than those routinely determined from ground-based astronomical observatories. A further star catalogue, the Tycho Star Catalogue of more than a million stars, is being compiled from additional data accumulated by the satellite. These catalogues will be of enormous value in astronomers' attempts to understand and describe the properties and evolution of stars, and the dynamical motion of these stars within our galaxy. In the process, HIPPARCOS has discovered many thousands of new binary star systems, measured the precise light variations of many hundreds of thousands of stars over its operational lifetime, and has provided an accurate and independent validation of the predictions of General Relativity.

Final accuracies on the stellar positions, distances and annual motions measured by HIPPARCOS, are in the range 1–2 thousandths of a second of arc, equivalent to the angular size of a golf ball viewed from the other side of the Atlantic Ocean. Distances of stars many millions of times further away than our own Sun are being measured, and their velocities are being measured with accuracies of a few hundred metres per second.

At the heart of the satellite was an extremely accurate optical telescope, which measured the separations between stars in different parts of the sky as the satellite turned slowly around its axis. The telescope mirrors on HIPPARCOS were so accurately polished that, if scaled up to the size of the Atlantic Ocean, deviations of the surface would nowhere exceed 10 cm in height. The global positional accuracies determined by HIPPARCOS are impossible to achieve from the Earth's surface, where the atmosphere, the temperature variations, instrument flexure, and the absence of all-sky visibility provide insurmountable barriers to ground-based observers.

Proposals have already been submitted to ESA to follow up its successful breakthrough into space astrometry with new missions proposed for launch early in the next millenium.